JPH0752391B2 - Translation method of source program including multiple database descriptions - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a translation method of a source program including a plurality of database descriptions, and more particularly to a method of expanding database record definition information of a source program including a plurality of database descriptions.

[Conventional technology]

Conventionally, in the translation method of the source program including this kind of database description, the database record definition information of the intermediate text expanded by the database description analysis means is fixedly expanded by the name tag of the record defined by the subschema object. Was there. Here, the sub-schema object is an object generated by compiling a program described in the sub-schema language.

[Problems to be Solved by the Invention]

In the above-mentioned translation method of the source program including the database description, the database record definition information is fixedly expanded by the name of the record defined in the subschema object. When multiple database descriptions are described, database record definition information with the same record name may be expanded, resulting in double definition of the record name and unique record name. It has the drawback that it cannot be referenced and a translation error occurs.

In view of the above points, an object of the present invention is to add a record name serial number, a subschema name serial number, and a program name serial number to a record label to change the record label so as to be unique. , It is to provide a translation method of a source program including a plurality of database descriptions capable of avoiding a translation error.

[Means for Solving the Problems]

A translation method of a source program including a plurality of database descriptions according to the present invention is defined by lexical decomposition means for inputting the source program and generating a first intermediate text which is lexically decomposed, and analyzing the database description in the first intermediate text. A database description analysis means for inputting a subschema object based on the subschema name and expanding the corresponding database record definition information into a second intermediate text, and this database description analysis means is activated by the definition unit of the subschema section Record information generating means for generating database record information consisting of all record names related to the name, presence / absence of use in a program, record name serial number, subschema name serial number, etc., and at the time of generating the database record definition information, From database description analysis means Database record name changing means for changing the name of the record to be unique by the record name serial number, the subschema name serial number and the program name serial number currently being processed in the database record information. 2 Data part syntactic and semantic analysis means for inputting an intermediate text and analyzing the data definition description to generate a name dictionary, and a record which is activated by the data part syntactic and semantic analysis means at the time of analysis of the data definition description and the level indicator is 01 Database record analysis means for searching the database record information for a name and registering in the name dictionary only the name of the record defined in the subschema object related to the record name that is used in the program; ,
The second intermediate text is input, and the user language token in the procedure division is replaced with the data name token in the name dictionary to generate a third intermediate text.

[Action]

In the method of translating a source program including a plurality of database descriptions of the present invention, the vocabulary decomposing means inputs the source program to generate a lexically decomposed first intermediate text, and the database description analyzing means analyzes the database description in the first intermediate text. The sub-schema object is input based on the parsed and defined sub-schema name, the corresponding database record definition information is expanded into the second intermediate text, and the record information generation means is activated from the database description analysis means to the definition unit of the sub-schema section. Generates database record information consisting of all record names related to the created subschema name, presence / absence of use in a program, record name serial number, subschema name serial number, etc., and database record name changing means defines the database record. The database is recorded when the information is generated. The name of the record is changed to be unique by the record name serial number, subschema name serial number, and the program name serial number currently being processed, which is started by the analysis means, and the data part syntactic and semantic analysis means The second intermediate text is input, the data definition description is analyzed to generate a name dictionary, and the database record analysis means is activated by the data part syntactic and semantic analysis means when the data definition description is analyzed, and the level indicator is 01.
The database record information is searched for the record name of, and only the name tag of the record defined in the subschema object related to the record name that is used in the program is registered in the name tag dictionary, and the attribute fetching means The second intermediate text is input and the third intermediate text is generated by replacing the user word token in the procedure division with the data name token in the name tag dictionary.

〔Example〕

 Next, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of a translation method of a source program including a plurality of database descriptions of the present invention. The translation method of the source program including a plurality of database descriptions of this embodiment is applied to a COBOL compiler, and the vocabulary decomposition means 1, the database description analysis means 2, the record information generation means 3, and the database record name change are performed. Means 4 and data part syntactic and semantic analysis means 5
, Database record analysis means 6, attribute acquisition means 7, source program 8, subschema object 9, first intermediate text 10, second intermediate text 11
, The database record information 12, the name tag dictionary 13, and the third intermediate text 14 constitute the main part.

Next, the operation of the translation method of the source program including the plurality of database descriptions of this embodiment configured as described above will be described. Note that a specific description will be given here with reference to FIGS. 2 to 4 as appropriate.

The vocabulary decomposition means 1 inputs a source program 8 including a plurality of programs in which the definitions of a plurality of subschema sections are described as a database description, that is, the definitions of the subschema sections are described, for example, as shown in FIG. , A first intermediate text 10 that is lexically decomposed into user words, reserved words, and the like is generated.

The database description analysis means 2 analyzes the database description in the first intermediate text 10 and includes record definitions of a plurality of subschema names as shown in FIG. 2 based on the defined subschema names. The generated sub-schema object 9 is input, and the record information generating means 3 and the database record name changing means 4 are activated to expand the corresponding database record definition information into the second intermediate text 11.

The record information generation means 3 determines whether or not the program is used for all record names related to the subschema name,
The database record information 12 including information necessary for making the name of the record unique is generated on the main memory (not shown).

Referring to FIG. 3, the database record information 12 is
It is composed of a database record name hash table, subschema name information, and record name information.

The database record name hash table is a hash table used to retrieve record name information associated with a subschema name.

The subschema name information is the subschema name defined in the database description of each program in the source program 8,
Subschema name serial number in source program 8, program name serial number in source program 8, number of record name information related to subschema name in subschema object 9, pointer to next subschema name information ( Next subschema record pointer), etc., and the number of subschema names defined in the database description in the source program 8 is generated.

The record name information is used when the record name in the subschema object 9, its record name length, whether the record name obtained as a result of analyzing the instructions in the program is used in the program, and the record name has the same hash value. It consists of a hash pointer (record hash pointer) that indicates the address of the next record information, a record name serial number in the program, etc., and is equal to the number of all record names related to the subschema name in the subschema object 9. Is generated.

The database record name mark changing means 4 generates a record name serial number, a subschema name serial number in the database record information 12 and a current process when generating database record definition information such as database control information area definition information and database record area definition information. Change the name of the record to be unique by the program name serial number inside.

For example, referring to FIG. 4, the record name a1 pointer in the database record definition information defined by the program a is the record name a1-01 according to the corresponding record name information and subschema name information in the database record information 12. Changed to -01-01. Also, program b
The record name b1 pointer in the database record definition information defined in 1. is changed to the record name b1-01-02-01, and the record name b2 pointer is changed to the record name b2-02-02-02.

Therefore, the record definition with the subschema name b has the same name as the record definition with the subschema name a (for example, a1-b).
Even if there is a record with 1), the record name serial number,
By adding the subschema name serial number and the program name serial number, the name tag of the record can be made unique.

The data part syntactic / semantic analysis means 5 inputs the second intermediate text 11, analyzes the data definition description, and extracts the name tag dictionary.
Generates 13.

The database record analysis means 6 is started by the data part syntax / means analysis means 5 at the time of analysis of the data definition description, the level indicator searches the database record information 12 for the record name of 01, and it is used by the program. Only the name tag of the record defined by the sub-schema object 9 related to the existing record name is registered in the name tag dictionary 13.

The attribute fetching means 7 inputs the second intermediate text 11 and replaces the user language token in the procedure division with the data name token in the name tag dictionary 13 to obtain the third intermediate text.
Produces 14.

〔The invention's effect〕

As described above, according to the present invention, when the database record definition information is generated, the record name serial number, the subschema name serial number, and the program name serial number are recorded based on the subschema name information and the record name information in the database record information. By adding after the name tag to change the name of the record so that it is unique, multiple database record definition information with different subschema names can be expanded in one source program and uniquely referenced. There is an effect that a program for processing a plurality of databases having different subschema names can be described in one source program without the risk of causing an error.

[Brief description of drawings]

FIG. 1 is a block diagram showing the construction of an embodiment of a translation method of a source program including a plurality of database descriptions of the present invention, and FIG. 2 is a database description and subschema objects in the source program shown in FIG. FIG. 3 shows an example of the database record information in FIG. 1, FIG. 3 shows an example of the database record information in FIG. 1, and FIG. 4 shows an example of the database record definition information in the second intermediate text in FIG. It is a figure. In the figure, 1 ... lexical decomposition means, 2 ... database description analysis means, 3 ... record information generation means, 4 ... database record name change means, 5 ... data part syntactic / means analysis means, 6 ... database Record analysis means, 7 ... Attribute fetching means, 8 ... Source program, 9 ... Subschema object, 10 ... First intermediate text, 11 ... Second intermediate text, 12 ... Database record information, 13 ... Nameplate dictionary, 14 ... 3rd intermediate text.

Claims (1)

[Claims] 1. A lexical decomposition means for inputting a source program to generate a lexically decomposed first intermediate text, and a subschema object based on a subschema name defined by analyzing a database description in the first intermediate text. And a corresponding database record definition information is expanded into the second intermediate text, and a database description analysis means is activated by the definition unit of the subschema section by the database description analysis means, and all record names and programs related to the subschema name. Record information generating means for generating database record information consisting of presence / absence of use in a database, record name serial number, subschema name serial number, etc .; Record in information Database record label changing means for changing the label of the record to be unique by the serial number, the subschema name serial number, and the program name serial number currently being processed; and the second intermediate text to input the data definition description A data part syntactic and semantic analysis means for parsing and generating a name tag dictionary; and searching the database record information for a record name whose level directive is 01, which is activated by the data part syntactic and semantic analysis means during analysis of the data definition description Database record analysis means for registering in the name dictionary dictionary only for the name of the record defined in the subschema object related to the record name that is used by the program, and the procedure for inputting the second intermediate text For the user language tokens in the section, A method of translating a source program including a plurality of database descriptions, comprising: attribute fetching means for generating a third intermediate text replaced with a data name token in Nari.